“My first flight in Nellie B was such an amazing experience, I found myself near as damnit laughing as I was climbing away from the aerodrome”, says Richard Grace. We’re sat in Air Leasing Ltd’s blister hangar at Sywell, Northants, on an icily cold late winter afternoon, and Richard is animatedly recounting that first sortie in Fighter Aviation Engineering Ltd’s Republic P-47D Thunderbolt. “It was a glorious day back then; early May 2018, the aircraft’s second flight since returning to the UK, wall-to-wall blue, pre-7am start and no one around. I was only up for ten minutes for some general handling, but had such a good time and felt so at home immediately – it’s like a warbird in slow motion!”
His enthusiasm for the ‘Jug’ is palpable, almost infectious, born out of a childhood spent amongst the airshow fraternity at Duxford whilst his mother, Carolyn, displayed the family Spitfire. “I grew up around the aircraft when it was with The Fighter Collection as No Guts, No Glory and watched it do some wonderful displays. I always loved it – it’s just cool, isn’t it?”, he continues. “I remember knocking around at a Duxford airshow the day after a big birthday ‘do’ in Cambridge, suffering from my very first hangover and not doing particularly much. Some of the Fighter Collection chaps thought it would be funny to have me pull the chocks on the Thunderbolt as it was sat there whomping away before a display – that’ll clear your head, for sure! Once I got into flying myself, I had no expectation that I would ever fly an aircraft like the P-47 and even once I got onto the bigger warbirds, I always thought it would be an aeroplane that would elude me as Thunderbolts are so rare. That first flight definitely gave me a genuine feeling of elation – airborne in it finally after all these years.”
P-47 Thunderbolt G-THUN, a former flagship of The Fighter Collection, departed the UK for a new life in the USA in 2007, its sale heavily felt by those who flew it and those who enjoyed seeing it fly alike. In subsequent years Richard set up the burgeoning Air Leasing facility at Sywell, restoring, maintaining and operating a raft of vintage aeroplanes for third party owners. As of early 2019, he has around 500 hours on high-performance piston warbirds, including the Sea Fury, Mustang, Buchón, Spitfire and Hurricane. The Thunderbolt’s acquisition by Fighter Aviation Engineering Ltd heralded its much-anticipated return to the UK, and shipment to the UK in a 40-foot container followed in spring 2018. Air Leasing reassembled the aeroplane at Sywell and 12 days after arrival it was rolled out, resplendent in a new paint scheme representing F4-J Nellie (updated in winter 2019 to Nellie B on the basis of new photographic evidence) of the 492nd Fighter Squadron, 48th Fighter Group, 9th Air Force. The P-47 flew again on 4 May, with Pete Kynsey at the controls.
Richard’s first flight followed shortly thereafter and he quickly built up time on type, his high-energy aerobatic air displays over summer 2018 belying the five ton Thunderbolt’s size. “When it’s parked up with gear and flaps down it looks like an ungainly brute,” he says, “but with everything cleaned up, it’s just the sleekest thing. To get it in the paint booth we had to take the windscreen and canopy off. I was looking out of the office window at it and was struck by just how sleek it is; perhaps that’s part of the reason I fly it the way I do!”
From the pilot’s perspective, the most involved element of any Thunderbolt pre-flight is manually turning the four-blade Hamilton Standard propeller through 18 blades to redistribute settled oil throughout the 18 cylinder Pratt & Whitney R-2800-51M1 engine (that powerplant coming from a C-46 Commando and having marginally different engine parameters to a stock Thunderbolt’s R-2800) before start-up; a strenuous task for pilot and ground crew, particularly on a hot day. With the 28 gallon oil tank positioned above the engine and lacking a shut-off valve, gravity will draw oil down into the engine over time, settling in the intake pipes or the combustion chambers of the lower cylinders. As oil is incompressible, and its pooling in the lowermost cylinders can stop the respective piston on the compression stroke as the engine fires, a process colloquially known as hydraulic locking, potentially causing critical damage to the respective cylinder or connecting rod. Whilst in the USA, Nellie B was mated with a radial oil clean kit which incorporated an electric shut-off valve for the oil tank, with a scavenger pump cycling oil from the engine back into the tank. That modification, Richard says, vastly reduces the possibility of hydraulic locking.
“If you’d gone from a Spitfire to a Thunderbolt, as the US Eagle Squadron pilots did in the Second World War, climbing into the P-47’s cockpit for the first time would feel like a big step up”, Richard says. The Thunderbolt’s pilot sits high in the broad cockpit at a “surprisingly upright angle”, he says. “It feels like sitting on a kitchen chair!” The cockpit layout itself is fairly standard, with a few idiosyncrasies specific to the P-47. Flight instruments are clustered centrally on the panel, with engine instrumentation sat on the right, the manifold pressure and tachometer positioned most prominently. Cylinder head, carburettor and oil temperature gauges, and the oil and fuel pressure indicators are clustered furthest right. Positioned alongside the pilot’s left thigh is a panel containing the numerous electrical switches and circuit breakers. A quadrant to their immediate left contains the throttle, propeller pitch and mixture levers. Aft of this sit the landing gear and flap control levers, each set at an angle of 90 degrees from each other to avoid in-flight confusion, the fuel tank selector, and the trim controls. To the pilot’s right is the tail wheel lock.
“The P-47 was designed in the 1930s, and it doesn’t quite have the prowess of the 1940s aircraft like the Mustang”, appraises Richard. “They hadn’t quite reached that point where instrumentation and systems are located logically and designed with ergonomics in mind. The throttle quadrant itself is badly laid out; the propeller and mixture control levers are side-by-side and stiff, and need some concentration to manipulate them. The throttle itself is huge, and sits almost in front of your face. It has a big throw from closed to full open, which also takes some getting used to. That said, the bubble canopy offers an excellent view and it has some interesting quirks, like the ability to drop the rudder pedals and cushion your heels on the cockpit floor. It’s a comfortable environment for long transits, much more so than the likes of the Spitfire and Hurricane, but there are some pretty serious ergonomic issues that were addressed by the time the Mustang came around.”
Prior to start-up, the oil cooler shutters and the cowl flaps are opened, with the throttle cracked by around one inch. Working forward from the rear port side of the cockpit, the undercarriage lever is checked in the Down position and the wing flap control handle moved full forward; the hydraulic system, pressurised by an engine-driven pump, actuates the wing flaps, cowl flaps and landing gear, and accordingly the wing flaps will raise once the engine has started. The fuel tank selector is then checked, and should already be set to the main tank – the Thunderbolt is only flown with its main tank fuelled (to a maximum capacity of 1,066 litres), and the fuel is usually left on to keep the carburettor lubricated to avoid the seals drying out. Moving forward to the throttle quadrant, the propeller control is set full forward and the mixture checked to be in Cut Off. The generator and fuel booster pump are both switched on, and fuel pressure monitored.
Starting the engine requires a degree of coaxing typical of radial engines: “The starter and primer switches are in an inconvenient spot low down to your left, but are well positioned in that you can push the start switch up with the middle of your finger and blip the primer with the tip of your finger. You can’t pre-prime because you’re just priming the carb, so you give it three bursts of prime for three seconds with your left hand, while your right hand sits on the mixture lever. Once you’ve cranked about eight blades, switch the magnetos on and it should start pretty quickly compared to some radials. Then you just move the mixture to Auto Lean and it’ll get going. It’s brilliant when it starts as there’s smoke everywhere, the obligatory oil spray onto the windscreen almost immediately, and the instrument panel is shaking all over the place as the pots disagree with one another until they agree to run as they’re supposed to and things smooth out. It’s got so much character, as the radials often do.”
The gradual engine warm-up can be protracted even in hot weather, taking up to ten minutes to achieve the desired cylinder head temperature (CHT) of 150°C, and an oil temperature of 40°C to avoid over-pressuring the engine on the initial taxi. Oil pressure settles to around 90psi and should sit comfortably there. Chocks removed by the ground crew, careful power input sees the aeroplane setting off to the runway holding point, a tap of the brakes confirming the system’s functionality. “There’s about 818kg on the tail wheel in the tail-down attitude”, Richard says. “It will hop on a brake check quite comfortably, and whilst it’d be a battle to get it up on its nose, it’s not beyond the realms of possibility and you keep that at the back of your mind. My hot tip is that unlike the Mustang, the Thunderbolt’s tail wheel is either locked or unlocked and you need to exercise caution when taxying. With it unlocked, it really is feral. It’s hard to explain how bad it is – if you were to taxi at high speed with it unlocked, it could get wild and it could get away from you pretty quickly. On long, straight runs I tend to lock the tail wheel as the forward view is pretty good.”
At the hold point, Richard will position the aircraft downwind or side-on to the prevailing wind for the power checks; this avoids air cooling the engine and allows the cylinders to sufficiently warm up. Moving the mixture into Auto Rich, he will slowly advance the throttle to 1,800 rpm and check the magnetos, looking for a drop of no more than 100 rpm. “The engine is so far away and on such a smooth mount that it’s very hard to hear a mag drop”, he adds. “It’s unusual in that the rpm gauge will be the only way of knowing whether you’ve lost a magneto.” Richard will then cycle the propeller control lever from full forward (fine) to full back (coarse) three times to run hot oil around the propeller components and improve their responsiveness.
“Personally, I leave the throttle up for a little longer to burn out any oil in the induction and throughout the inlet and exhausts, to avoid seeing puffs of white smoke popping out of the engine on take-off, which can be a bit disconcerting”, he says. “I’ll then bring the power very carefully back to idle. You do everything super slowly in these aeroplanes. In the P-47, there’s a whomping great propeller up front; if the engine is turning the propeller and you chop the power, for example, the propeller would start turning the engine and you’d change the loading on all of the internal components of that engine. That would upset things very rapidly, and you’re at risk of causing serious damage to that engine.”
He continues: “I check that idle power will give me a sensible rpm that will permit the aeroplane to land within the distance I want it to land – if you haven’t closed the throttle to idle on landing in these aeroplanes, they will just float the length of the runway. Even if the tacho’ is showing 1,000 rpm on landing, it’s still kicking out some thrust and will absolutely keep the aeroplane airborne, and that’s somewhere you really don’t want to be at slow speed, close to the ground and running out of runway. It’s important to have that idle rpm in mind.”
Throttle is then advanced to 1,000 rpm to avoid fouling the spark plugs whilst the final instrument checks are carried out, and the cowl flaps and oil cooler door opened half way. “At that point I put my right hand on the tail wheel lock,” says Richard, “because I don’t want to forget that. It taxies well at 1,000 rpm so I can leave the throttle and have my left hand on the stick with the other on the tail wheel lock until I’m lined up on the runway. You then increase the throttle to 30inHg (“) and release the brakes as you continue to increase the power. The second it lurches forward, it feels like it slows down! You advance the power to 50″, checking the temperatures and pressures intermittently, and the airspeed and rpm will slowly build it until the aeroplane flies itself off the ground”. He describes the take-off run as a docile affair, largely due to the aeroplane’s sheer mass and the enormous gear track (around 18ft) that precludes the need for torque-counteracting rudder or opposite aileron input.
“As the power is set and I’m rolling, I’ll move my hand to the gear selector and once I’m satisfied it’s airborne, I’ll pull the wheels up within around ten seconds”, Richard explains. The gear is retracted by depressing the button at the end of the selector lever with the left thumb and pushing down to release the lock before raising. “At the vast majority of airfields in the UK, the runway is of a length that if you lose the engine on take-off, you won’t be able to get back on the ground and stop the aeroplane before running out of runway. You’ll end up going through a hedge and into a field, which will take the gear off and you’ll likely come to grief too. I know what it takes to repair an aeroplane that has had its gear torn off, and if that happens, it’ll be on the ground for a very long time. Even if the gear is transiting as the engine quits and you end up in a field, you’ll have lessened the damage. That’s why I get the gear up immediately – it’s risk mitigation.
“After my first flight, I described the Thunderbolt as climbing like a shot pigeon”, Richard laughs. “As you clear the runway you think, okay, let’s go up. But the second you bring the nose up the airspeed starts to wash off and you think, ah, this is novel – I guess we aren’t going anywhere! It tracks along so fast that it’s actually climbing quite well, but at a shallow angle as compared to any other warbird. We flew a formation take-off with the Thunderbolt and Mustang last year. Jon Gowdy was in the P-47 and I was in TF-51 Contrary Mary. I thought, I’ll build up power on the brakes and release as soon as the Thunderbolt starts to roll. I expected I would need to come back on the power, and would end up getting airborne before the P-47. It was the polar opposite! The second he let his brakes off, I released mine and it was like nothing happened by comparison – I was sat watching the P-47 disappearing off down the runway! I put on as much power as you’d want to put on in a Mustang, and he was airborne before me. The Mustang cleans up faster than the Thunderbolt, so I closed the gap as the undercarriage was retracting on both aeroplanes, but the second the wheels were up the P-47 just went. It was a good 10kts faster and I eventually had to ask him to come back on the power as I was giving it everything and couldn’t catch up!”
The gentle climb at 42″ and 2,400rpm is a pedestrian affair. Once in the cruise, the cowl gills are closed (via a choke lever on the bottom right of the instrument panel) to reduce drag and avoid excessive cooling. With power settings of 35-40″ and 2,000 to 2,200 rpm, the P-47 should cruise at 210kts – “sprightly for a biggun!” – with the CHT sat at 150-180°C, oil temperature hovering around 70°C (regulated by operation of the oil cooler shutters via a switch on the main instrument panel) and the oil pressure pegged at 90psi. The physical air temperature in the cockpit is very high at all times of the year, the air vents doing nothing but “blowing hairdryer hot air into your face”. The Mustang, by comparison, takes cool air from the radiator scoop and pumps it up through the back of the canopy and down onto the pilot’s face. “No such luxury in the Thunderbolt!”, laments Richard.
“The hot tip for high-performance manoeuvres in the P-47 is that you can’t be too ambitious with it”, he says. “You’ve got to be a realist about what it’s going to achieve, and consider when you’re going to achieve that in the display. You’ve got to save your energy through smooth pulls; once you’ve really slowed it down, it’ll take a lot of work to get its airspeed back up. If you’re midway through a display sequence and only have the airspeed for one final half-cuban, it’s better to fly a steep wingover to win yourself energy for another three or four vertical manoeuvres.”
Richard continues: “Aerobatically, there’s a sweet spot in all of these aeroplanes. I spend a lot of my time initially on a new type at height, just descending to the airspeed I want to be entering aerobatics at and pulling up into the vertical, checking the airspeed in the vertical and seeing how fast I can get it going in the climb. If I enter at the same airspeed and find a higher airspeed in the climb, it means I got the g loading right at the bottom. Then I know where the sweet spot is for a relevant speed. I’ll know that if I’m slower, I’ll need to pull slightly more to stay in that sweet spot. You have to spend a lot of time flying these figures to optimise the performance of an aeroplane. The key to happiness is finding the most speed economic way to get the aeroplane to the vertical.”
Positioning the Thunderbolt for an aerobatic display practice at Duxford, the aeroplane’s summer base, Richard will climb to 4,500ft (just shy of Stansted airspace), maintain climb/aerobatic power of 42″ and 2,400 rpm and fly large circles until the aeroplane is going as fast as it can achieve in level flight at that power setting. “I’ll then start a very steep descent towards the airfield”, he says. “In a Spitfire, that would be near vertical – in the Thunderbolt I’m lessening that angle slightly due to the weight of the thing, but the objective is to fly the aeroplane as fast as it will let me go.” That has the P-47 screaming over the airfield at low-level and at least 320kts (368mph) for the first pass of the routine.
The entry to the first manoeuvre – an 80 degree vertical climb and half roll to complete a reverse cuban – is where that “sweet spot” first comes into play. The mass and chord of the Thunderbolt’s wing are such that a hard pull into the vertical will cause the aeroplane to ‘mush’, with the induced drag sapping the airspeed. “If you jar the P-47 any way and, say, knick that first g too quickly as you yank it into a 2g climb, you’ve probably just eaten 10kts”, Richard cautions. “The key is to ever so slowly apply the elevator to creep the accelerometer up, never pulling more than 4g as an absolute maximum. More so than most warbirds, the Thunderbolt does not appreciate high g loading, and it’ll bite you hard if you’re too aggressive with the control inputs.”
On the climb into the vertical, Richard will be closely monitoring the oil temperature and pressure; a habit he taught himself early on, and one honed when flying the Sea Fury, where oil pressure is checked every 30 seconds. “It’s good practice to monitor engine parameters in the climb rather than staring blankly at the altimeter”, he says. “If the engine’s going to fail, I want to know at the top of the manoeuvre to give myself maximum altitude to work with.” In the climb, gyroscopic forces react to the right-turning propeller by yawing the aeroplane to the left; that gyroscopic precession is particularly pronounced in Second World War fighters with high propeller rpm. “You’ll instinctively be applying rudder as you enter the manoeuvre to counteract the torque and avoid the yaw washing the airspeed off, easing the pressure off until your feet are both central in the early stages of the vertical pull before feeding in opposite rudder to keep the aeroplane in balance as the airspeed decreases. It is a permanent battle in the high-performance warbirds, and if you left your feet in the centre it’d be pretty exciting. In the Mk.XIV for example, it’s laughable. Loads of rudder pressure on the way in, then through the vertical it’s almost an immediate change to the other foot and at the apex of the vertical figure, you’ve got full rudder on just to barely keep it in a straight line.
“I construct my vertical display routine around gate airspeeds”, Richard explains. “I know what airspeed I entered the figure at, and I know that if I’ve applied the right amount of elevator on the pull-up, the Thunderbolt will be at 100-110kts at the apex of the loop. Regardless of the entry airspeed, I’ll aim to keep that over-the-top airspeed consistent by adjusting the amount of elevator input in the first half of the figure. I always start with a vertical climb and a half-roll, followed by a loop; what happens next depends on the type. I know that at my entry airspeed, the reverse cuban will give me enough altitude to recover, and it’s the loop that’s key. In the P-47 I’ll only loop early on in the display when the airspeed and energy are at a premium. For that reason, I’m usually only checking my gate entry airspeeds for the first two figures, as that’s where I need to exercise caution.
“Airspeed is key in aeroplanes of this ilk”, he stresses. “It’s the final quarter of the loop that really matters, and your altitude and airspeed depend entirely on how you’ve handled the aeroplane to that point. The primary concern is that you might end up too fast at the apex. If you’re at 140kts going over the top, which is completely achievable in this type of aeroplane, that means you’ve pulled too hard in the first and second quarters of the loop. If you’re at 140kts at the apex you’re probably 300ft lower than you want to be, but your airspeed will increase during the final half of the loop such that you’ll be around 1,000ft lower than you entered the loop at. If you use gate heights only, you might achieve the altitude you were expecting to get, but what if you entered the loop 10kts faster than you thought you did? You could check your altitude at the apex and think, ‘happy days’, but if you haven’t checked the airspeed you might be doing 140kts and you might be about to recover 1,000ft lower than you entered at. The first time I ever looped a Spitfire, I was in ML407 with Pete Kynsey in the front and I pulled up, thought it was going pretty well, checked the altitude at the top of the loop and was happy with what I saw. Pete suddenly called out, ‘No, no, don’t pull through – check the airspeed’. I was at 140kts, and turned it into a half-cuban.”
As the airspeed and energy bleed off, vertical manoeuvres give way to rolling figures and topside passes. The ailerons are very good for an aeroplane of its size, and need little application force – around half of the stick’s travel – to initiate a rapid rate of roll. Breaking into the circuit after a sortie and pulling the power back (reducing rpm before manifold pressure to avoid the propeller driving the engine), the airspeed should decay fairly rapidly. A high gear and flap limit of 170kts allows early onset of drag in the downwind, albeit Richard cautions that the Thunderbolt will not maintain level flight with the gear and flaps down in a turn, even at full power. “If you’re turning into final, you can pile the coals on all the way to full power and whilst you’re lessening the rate of descent, you will still be descending”, he says. “That could put you in a difficult position if you weren’t awake to that fact.
“If you’re turning final at 120kts, you’ll be coming over the hedge at around 100kts and you really won’t have a lot of power left. If you’re below 100kts on short final, you have to look out of the window and fly by feel alone, as the airspeed indicator won’t be of any use. Generally you’ll two-point it in a tail-down attitude, and once you chop the power to idle you know it’ll land and stay on the ground. Those huge undercarriage legs absorb the landing and there’s little risk of floating or bouncing. As with the take-off, the large gear track and mass of the thing mean it’ll sit pretty in the middle of the runway with little risk of deviations. It’s a docile affair, you just need to be conscious of the conflict between rpm and manifold pressure, and the aeroplane’s inability to climb out of a level turn with gear and flaps down.” After sitting on the ground for a few minutes to allow the CHT and oil temperature to stabilise, the engine is shut down by cutting the mixture, with the magnetos and electrics then switched off in turn.
“You hear a lot about how the Thunderbolt flies like a brick, and is a bit of a freight train to handle”, Grace reflects. “Yes, it’s an enormous aeroplane and that could be daunting when you first approach it, but it’s so manoeuvrable. It’s every inch the fighter. With a light fuel load you could loop from 250kts, which is unheard of for an aeroplane of its size and weight. All these aeroplanes perform so differently, you just need to sit down and figure it out in your head, then have the opportunity to practice and really learn the aeroplane. It’s an awesome machine and down at low-level and 320kts you really get a sense of why it was such a successful ground attack aeroplane. We look after a number of ex-Fighter Collection aeroplanes and interestingly, this one came to us with more post-restoration hours on it than any of the others, which is perhaps indicative of how nice it is to fly.
“No matter how many flights I’ve had in it and how many displays I’ve flown, I’ll always come back to that very first flight from Sywell. Genuine elation.”
With thanks to Richard Grace and Fighter Aviation Engineering Ltd. Flights alongside the P-47 Thunderbolt, and passenger rides in Second World War fighters, are available through Ultimate Warbird Flights at Sywell.